Strength of Stacking Technique of Ensemble Learning in Rockburst Prediction with Imbalanced Data: Comparison of Eight Single and Ensemble Models

Rockburst is a common dynamic geological hazard, severely restricting the development and utilization of underground space and resources. As the depth of excavation and mining increases, rockburst tends to occur frequently. Hence, it is necessary to carry out a study on rockburst prediction. Due to the nonlinear relationship between rockburst and its influencing factors, artificial intelligence was introduced. However, the collected data were typically imbalanced. Single algorithms trained by such data have low recognition for minority classes. In order to handle the problem, this paper employed stacking technique of ensemble learning to establish rockburst prediction models. In total, 246 sets of data were collected. In the preprocessing stage, three data mining techniques including principal component analysis, local outlier factor and expectation maximization algorithm were used for dimension reduction, outlier detection and outlier substitution, respectively. Then, the pre-processed data were split into a training set (75%) and a test set (25%) with stratified sampling. Based on the four classical single intelligent algorithms, namely k-nearest neighbors (KNN), support vector machine (SVM), deep neural network (DNN) and recurrent neural network (RNN), four ensemble models (KNN–RNN, SVM–RNN, DNN–RNN and KNN–SVM–DNN–RNN) were built by stacking technique of ensemble learning. The prediction performance of eight models was evaluated, and the differences between single models and ensemble models were analyzed. Additionally, a sensitivity analysis was conducted, revealing the importance of input variables on the models. Finally, the impact of class imbalance on the prediction accuracy and fitting effect of models was quantitatively discussed. The results showed that stacking technique of ensemble learning provides a new and promising way for rockburst prediction, which exhibits unique advantages especially when using imbalanced data.

     

Establishment and characterization of a testicular Sertoli cell line from olive flounder <Emphasis Type="Italic">Paralichthys olivaceus</Emphasis>

The culture of Sertoli cells has become an indispensable resource in studying spermatogenesis. A new Sertoli cell line (POSC) that consisted predominantly of fibroblast-like cells was derived from the testis of the olive flounder Paralichthys olivaceus and sub-cultured for 48 passages. Analysis of the mtDNA COI gene partial sequence confirmed that the cell line was from P. olivaceus. Cells were optimally maintained at 25°C in DMEM/F12 medium supplemented with fetal bovine serum, basic fibroblast growth factor, and epidermal growth factor. The growth curve of POSC showed a typical “S” shape. Chromosome analysis revealed that the cell line possessed the normal P. olivaceus diploid karyotype of 2n=48t. POSC expressed dmrt1 but not vasa, which was detected using RT-PCR and sequencing. Immunocytochemistry revealed that the cells exhibited the testicular Sertoli cell marker FasL. Therefore, POSC appeared to consist of testicular Sertoli cells. Bright fluorescent signals were observed after the cells were transfected with pEGFP-N3 plasmid, with the transfection efficiency reaching 10%. This research not only offers an ideal model for further gene expression and regulation studies on P. olivaceus, but also serves as valuable material in studying fish spermatogenesis, Sertoli cell-germ cell interactions, and the mechanism of growth and development of testis.     

利用MODIS资料遥感水体上空气溶胶粒子尺度的数值试验

利用卫星探测辐射的光谱依赖性来反演大气气溶胶粒子尺度分布,首先要建立反演算法并检验其可行性。本文参照文献[1]对暗的海洋上空MODIS资料的处理方法,针对大陆水库上空卫星探测MODIS辐射信号,利用6S(Second Simulation of the Satellite Signal in the Solar Spectrum)辐射模式建立一个查算表(Look-up table:LUT)来反演大气气溶胶粒子尺度分布;然后通过数值试验对所建立的查算表进行检验,讨论其方法的可行性。数值试验结果表明,查算表对大陆性气溶胶尺度分布有反演能力,但拟合误差sεl值随着气溶胶光学厚度的增加而增加。     

Study of Influence of Regional Geodynamic Background on Strata Behaviors in China's Tongxin Mine

Strata behaviors are mainly affected by regional geodynamic background. The influence of rock mass stress and energy distribution on strata behaviors in the Tongxin mine is studied in terms of regional tectonic movement, seismic activity and tectonic stress field. The results show that the extrusion lifting movement of Kouquan fault adjacent to the Tongxin mine results in the stress concentration in the rock of the Carboniferous coal bed and accumulation of a large amount of elastic energy and forms structural background of Tongxin mine. Due to various seismic activities in the mine area, the strain energy is known to reach much higher levels, up to 0.5×10⁸ J^1/2. Since the stratigraphic structure is sensitive to the mining operation, the strain energy could cause strong strata behaviors. A special geological structure model of the Tongxin mine is established based on the geodynamic division method. The distribution of regional structure stress field is determined by the rock mass stress analysis system. Based on this model, Tongxin mine is divided into five areas with high stress, eight areas with low stress and eight areas with gradient stress. The strong strata behaviors mostly occur in high stress areas. These results could provide guidance to predict the strength of regional or mine pressure and control strata behavior in different areas.     

一次三峡大暴雨的地转螺旋度分析

本文引入了地转螺旋度概念，并将其用于一次大暴雨的分析，结果发现；在地转螺旋度垂直剖面图上，在雨区上空３００ｈＰａ附近有一个负的极小值中心，在５００ｈＰａ有一个正的极大值中心，该分布形成原因可用图６解释。     

广东省海岸表层沉积物的重矿物特征及其与沉积环境的关系

广东省海岸表层沉积物中的重矿物近40种。含量较高的有钛铁矿、赤矿铁、磁铁矿、话石、电气石、黑云母和普通角闪石等。自生矿物主要为硬石膏、自然铜、黄铁矿及菱铁矿。这些矿物来自附近的母岩。根据物源性质不同,可将不同岸段重矿物区分为陆源矿物组合、近源-自生矿物组合和近源矿物组合三种:除物源控制外,重矿物含量、种类、形态特征及其在平面上的分布主要受沉积环境影响。据此,可将重矿物分布类型划分为河口型、海湾型和海岛型三类,且与上述不同矿物组合性质相对应。     

对GPS接收机测试评估的探讨

随着GPS的广泛应用和GPS产业的迅速发展，GPS厂家众多，生产的接收机类型更是数以百计，整个GPS接收机市场可以说是鱼目混珠，识别困难；大多数购买者只能听信厂家的技术指标，如何对GPS接收机的性能进行测试评估的意义显得很重要，本文基于此对GPS接收机测试评估展开探讨和研究。     

滆湖磷迁移过程的研究(一)——磷迁移模型的建立

通过滆湖磷的来源途径(入湖河道、湖区径流、湖面沉降、养殖技饵、底泥释放)和湖体各要素(水体、浮游植物、大型水生植物、鱼类等)中磷迁移过程调查资料的分析,建立了描述磷在上述各要素中迁移过程的数学模型.经实测资料验证,模型的计算值与实测值的平均相对误差在9.9%-18.6%.基本反映了磷的迁移过程及动态变化规律,对该湖磷浓度预测及磷资源合理利用具有重要的意义.     

CMIP3与CMIP5模式对中国西北干旱区气温和降水的模拟能力比较

通过对15组CMIP3和CMIP5两代模式集合平均对中国西北干旱区气温和降水的模拟能力比较,发现CMIP5模式对气温和降水的模拟更接近观测值。CMIP5模式模拟年、春季、夏季、秋季平均气温的相关系数比CMIP3模式分别提升了0.15、0.13、0.24和0.02,冬季下降了0.07。CMIP5模式对西北干旱区的平均气温变化趋势的模拟效果比CMIP3有所提高,对年、春季、夏季、秋季、冬季趋势的模拟偏差比CMIP3分别减少了0.03℃/10a、0.10℃/10a、0.01℃/10a、0.06℃/10a、0.14℃/10a。对西北干旱区平均气温年、季的模拟偏差分布上,CMIP5模式的偏差均比CMIP3低1~2℃。但是天山区年、季节平均气温的模拟与整体模拟偏低情况相反,CMIP3和CMIP5分别偏高3~6℃和1~4℃,对夏季的模拟偏高最严重,分别达到6℃和4℃。CMIP5模式整体对西北干旱区降水量的模拟结果与观测值的平均相关系数与CMIP3相差不大,均不超过0.1,而且偏差仍然较大。CMIP5模式对西北干旱区的降水量的变化趋势模拟效果比CMIP3有所降低,对年、春季、夏季、秋季、冬季趋势的模拟偏差比CMIP3增加了0.67 mm/10a、0.23 mm/10a、0.51 mm/10a、0.11 mm/10a、0.14 mm/10a。CMIP5模式对年、春季、夏季、秋季和冬季的降水量模拟的均方根误差相比CMIP3分别减少77.6 mm、25.5 mm、25.0 mm、18.8 mm和13.9 mm。在空间上,CMIP5模式对年、季节降水模拟仍然偏高,但是比CMIP3有明显缓解;CMIP3和CMIP5模式对夏季天山区年降水量和夏季降水量的模拟也与大部分区域偏高的趋势明显相反,两代模式对夏季天山区的降水模拟均偏低50 mm左右。     

Seismic fracture evaluation of diaphragm joints in welded beam-to-box column moment connections

Steel box columns are widely used in steel building structures in Taiwan due to their dual strong axes. To transfer the beam-end moment to the column, diaphragm plates of the same thickness and elevations as the beam flanges are usually welded inside the box column. The electro-slag welding (ESW) process is widely used to connect the diaphragms to the column flanges in Taiwan because of its convenience and efficiency. However, ESW may increase the hardness of the welds and heat-affected zones (HAZs), while reducing the Charpy-V notch strength in the HAZ. This situation can cause premature fracture of the diaphragm-to-column flange welds before a large plastic rotation is developed in the beam-to-box column joints. To quantify the critical eccentricity and the effectiveness of fracture prediction, this study uses fracture prediction models and finite element model (FEM) analysis to correlate the test results. In this study, two beam-to-box column connection subassembly tests are conducted with different loading protocols and ESW chamber shapes. To implement a fracture prediction model, the material parameters are established from circumferential notched tensile tests and FEM analysis. Test results indicate that the fracture instances can be predicted on the basis of the cumulative plastic deformation in the HAZs. Analytical results indicate that fracture instances and locations are sensitive to the relative locations of the ESW joints and beam flange. Tests also confirm that the possible fracture of the diaphragm-to-column flange joints can be mitigated by enlarging the chamber of the ESW joint.